Associations of body composition with type 2 diabetes mellitus.

The aims of this study were to establish the associations of stature, body mass index, waist to hip ratio, and waist circumference with Type 2 (non-insulin-dependent) diabetes mellitus in a random sample of 5887 men and 7018 women aged 20-59 years in a cross-sectional study set in The Netherlands. The crude prevalence of Type 2 diabetes (overall 1.58% in men, 0.94% in women) was significantly (p < 0.01) higher in shorter subjects and those with high body mass index, high waist to hip ratio, and larger waist circumference. Odds ratios and 95% confidence intervals (95% CI) were adjusted for age, cigarette smoking, alcohol consumption, physical activity, and education. Compared to the tallest tertile of height, odds ratios for Type 2 diabetes were 4.4 (95% CI: 1.3 to 11.5) in men and 1.6 (95% CI: 0.8 to 3.2) in women whose height was in the shortest tertile. Compared to the lowest tertile, odds ratios for Type 2 diabetes were 18.4 (95% CI: 4.3 to 78.5) in men and 5.3 (95% CI: 2.0 to 14.0) in women with waist to hip ratio in the highest tertile, 4.1 (95% CI: 2.0 to 8.4) in men and 2.1 (95% CI: 1.0 to 4.2) in women with body mass index in the highest tertile, and 4.9 (95% CI: 2.1 to 11.7) in men and 2.7 (95% CI: 1.2 to 5.9) in women with waist circumference in the highest tertile. In conclusion, although in longitudinal studies waist is a powerful predictor of diabetes incidence, Type 2 diabetes in a cross-sectional survey is associated with shortness in stature, as well as large waist circumference and high body mass index, and particularly strongly with high waist to hip ratio, suggesting that the development of Type 2 diabetes may modify hip circumference independently of body fat.